children, regardless of their number. This does not occur with those factors responsible for intelligence. The factors operating to impede intellectual development in children from large families, whatever they may be, are not compensated for even in upper-middle families.
In his 1966 Galton Lecture, Tanner (46) reviewed the evidence gathered to date of the association between height and intelligence, and between those two attributes and family size. He raises an interesting point with regard to findings concerning family size:
To summarize then: according to present data, children with many sibs in the house are retarded in their height growth from an early age compared with children of the same social class with few sibs. This is especially true of children in poorly-off families. They also score lower in tests of intelligence or attainment. By the time adulthood is reached they have not caught up in intelligence tests, but this may be the result only of the vicious circle in educational opportunity described above. . .. (46, p. 130)
The "vicious circle" mentioned by Tanner is worth describing for those inclined to wonder whether the actual differences in the I.Q.'s found in children of different family sizes are in fact of any real consequence. Tanner noted that a 9-point difference in I.Q. amounts to "two thirds of the standard deviation of the test score, and in the 11+ exam, for instance, corresponds to a difference of about 15 percentile ranks at the level usually used for pass or fail." Later on he notes that children who have that advantage in the 11+ exam "obtain an increasing educational advantage thereafter, simply as a result of passing these tests. Hence they would remain always ahead, an example of the classical self-fulfilling prophecy or positive feedback." The power of the "self-fulfilling prophecy" has been recently documented by Rosenthal and Jacobson (47), who showed convincingly that teacher expectations based on the reported test performance (the teachers were shown no actual results) of children have a marked effect on the performance of the children in the classroom situation.
The Effects on Parents
Children are more susceptible to environmental influences and the effects of these influences are more easily measured. Nevertheless, there is some evidence that family size takes its toll of parents, too.
Family Size and Parental Health. Chen and Cobb (27) cited three pertinent studies. One "... has shown a direct linear relationship between the frequency of peptic ulcer and number of children for a group of employed men." A study of women showed "... a positive association between rheumatoid arthritis and large numbers of children." A study of blood pressures, interestingly enough, has shown no such relationship: ". . .blood pressuresa family (41).